BUG: BooleanArray+pd.NA altering array inplace (#45421)

Author: jbrockmendel

doc/source/whatsnew/v1.5.0.rst

@@ -185,7 +185,7 @@ Timezones
 
 Numeric
 ^^^^^^^
--
+- Bug in operations with array-likes with ``dtype="boolean"`` and :attr:`NA` incorrectly altering the array in-place (:issue:`45421`)
 -
 
 Conversion

pandas/core/arrays/boolean.py

@@ -494,7 +494,8 @@ def _arith_method(self, other, op):
         if mask is None:
             mask = self._mask
             if other is libmissing.NA:
-                mask |= True
+                # GH#45421 don't alter inplace
+                mask = mask | True
         else:
             mask = self._mask | mask
 

pandas/tests/arrays/masked/test_arithmetic.py

@@ -47,16 +47,22 @@ def test_array_scalar_like_equivalence(data, all_arithmetic_operators):
         tm.assert_extension_array_equal(result, expected)
 
 
-def test_array_NA(data, all_arithmetic_operators, request):
+def test_array_NA(data, all_arithmetic_operators):
     data, _ = data
     op = tm.get_op_from_name(all_arithmetic_operators)
     check_skip(data, all_arithmetic_operators)
 
     scalar = pd.NA
     scalar_array = pd.array([pd.NA] * len(data), dtype=data.dtype)
 
+    mask = data._mask.copy()
     result = op(data, scalar)
+    # GH#45421 check op doesn't alter data._mask inplace
+    tm.assert_numpy_array_equal(mask, data._mask)
+
     expected = op(data, scalar_array)
+    tm.assert_numpy_array_equal(mask, data._mask)
+
     tm.assert_extension_array_equal(result, expected)